Meet The Disruptors: NEC’s Chris White On The Five Things You Need To Shake Up Your Industry

Understand how things scale. Having access to a disruptive technology is one thing. Deploying it at scale is very different. It’s not only knowing the problem and that we have a solution. It’s how do I actually deploy that solution at scale?

As a part of our series about business leaders who are shaking things up in their industry, I had the pleasure of interviewing Chris White.

Christopher A. White is the president of NEC Laboratories America, where he leads a team of world-class researchers focusing on diverse topics, from sensing to networking to machine learning-based understanding. Before NEC, he spent 22 years at Bell Labs, leading the Algorithms, Analytics, Augmented Intelligence, and Devices (AAAID) research lab.

He holds a BS in Chemistry from Carnegie Mellon University and a Ph.D. in theoretical quantum chemistry from the University of California in Berkeley. His research interests include creating computational models and methods for simulation, modeling, and controlling interesting physical and digital systems.

Thank you so much for doing this with us! Before we dig in, our readers would like to get to know you a bit more. Can you tell us a bit about your “backstory”? What led you to this particular career path?

My passion for computers took me to Carnegie Mellon. While I took computer courses in college, I majored in chemistry because I didn’t want to focus solely on writing database code, which was the major focus at the time.

I wanted to define my direction rather than coding things for other people, and computational chemistry was very big at the time.

While at graduate school at UC Berkeley, my focus was using computer science techniques to solve large-scale quantum chemistry problems. Quantum mechanics define the equations that provide exact answers, but those exact answers don’t always scale effectively with the size of the system treated.

The challenge is that it is easy to compute small three-atom molecular systems like water, but unfortunately, as you add more atoms to the system, the computation time grows by factors of ten.

I was building computational models to do those very large-scale kinds of systems.

After graduating from Berkeley, my goal was to leverage my passion for teaching to become a professor. The path I chose was a bit different than most. I explored and accepted an opportunity at Bell Labs, which was doing some innovative research with one foot in industry and the other in academia. It was an ideal place to spend five years in a more lucrative environment than as an assistant professor.

At Bell Labs, the flow of interesting and relevant problems was more extensive than in academia.

My experience at Bell Labs directly led to the opportunity at NEC Labs, which offered a constant flow of these types of interesting and real-world problems from customers that only can be found in the context of an industrial lab.

Can you tell our readers what it is about the work you’re doing that’s disruptive?

Many organizations have R&D labs, which typically try to improve the current product with an incremental 10–15% improvement. If they don’t continually improve, market share is slowly lost. This approach centers on stability and long-term survival.

As an industrial lab, here at NEC Labs America, our focus is centered on disruptive research. Opening new markets. Creating new products. Finding an entirely new way of doing something that no one has been able to do before. We don’t look for a 10% improvement. We look to improve performance by factors of 10.

My goal is to take problems we see from customers and not just solve those problems with specific solutions but to find general solutions that can be applied in many different contexts beyond the original specific scenario.

With optical fiber sensing, we combine research on optical sensing, optical localization, and data fusion to build broadly applicable applications. With machine learning, we’re exploring the intersection between physics and machine learning to predict what happens in the world.

We’re consistently looking for more problems to solve across industries with our solutions. When building these applications that sense the world, we do it in a context of limited resources. This is the concept of Elastic AI, which is how you ensure the amount of effort put towards solving a problem is proportional to the value of finding that solution, and that elastically can change as things go forward.

Can you share a story about the funniest mistake you made when you were first starting? Can you tell us what lesson you learned from that?

Too many mistakes to count! In many cases, mistakes have been the sources of significant advances.

One mistake, which is made very generally across the board, is the thought that the best technology wins. We’ve been trained to believe that great technology leads to great innovation.

The interesting aspect of being an industrial lab is that we constantly check our solutions against a customer need. We drive to develop a technology that best matches those customer needs, and therefore we don’t waste effort by developing a technology beyond satisfying that need.

In reality, it’s more important to understand the problem you’re trying to solve.

One specific example in my career focused on acoustic scattering, launching acoustic waves through a material to understand the characteristics of that material. Organizations, including semiconductor manufacturers, use this type of technology for device fabrication.

I received one of my first patents for this technology. Interestingly, it wasn’t a semiconductor manufacturer that first licensed it. It was a peanut butter company! We put in a lot of time and effort to build this technology for really high-tech applications, and in the end, it was a company that made peanut butter that used it. Apparently, the taste of peanut butter depends on the size of the particles that make up the peanut butter. And, of course, you can’t shine a light through most peanut butter. So, using this acoustic method satisfied the requirements for peanut butter, and therefore, it was a relevant solution.

We all need a little help along the journey. Who have been some of your mentors? Can you share a story about how they made an impact?

Over my career, there have been way too many amazing individuals who served as mentors. It’s impossible to call out just one or two.

One thing I can say is that the environment we all worked in at Bell Labs was critical to my career growth. It was an environment where no one felt they were as good as the person who sat two doors down. We all felt a bit of imposter syndrome because the people around us were all so amazing. It was great because it drove all of us to be better.

It ended up being a very collaborative environment. One day you might interact with an eventual Nobel Prize recipient. On the same team, there would be an intern just out of school. Everyone worked together to solve problems without a stiff hierarchical structure.

In today’s parlance, being disruptive is usually a positive adjective. But is disrupting always good? When do we say the converse that a system or structure has ‘withstood the test of time’? Can you articulate to our readers when disrupting an industry is positive and when disrupting an industry is ‘not so positive’? Can you share some examples of what you mean?

I’m not entirely sure that I agree that disruptive is usually a positive adjective. Certainly, in business, it is. But if I tell other people outside of business that our research is disruptive research, they don’t often think that it’s a positive thing. I must explain what I mean by that and how disruptive research is differentiated from incremental research.

There’s an interesting evolution that happens in many industries. We move between periods of rapid progress generated by repeatedly applying a simple set of rules.

Eventually, we create technology or solutions to solve a current set of problems. It starts with solving a hard problem and then applying that solution to many other problems. Over time, while we thought these additional problems were hard, by using the same solution, we’ve proven that they are equivalent to the first problem.

And once you’ve exhausted problems in that class, you need to find a new class of problems to solve. And that’s the disruptive piece, finding new opportunities and developing solutions to replicate repeatedly. As an industrial lab, our mandate is to find classes of problems where a specific solution to one problem can be applied to many problems in that same class.

Industrial Labs are always searching for where these disruptions are.

Can you please share five ideas one needs to shake up their industry?

Five concepts need to be understood to shake up an industry.

1 . Understand the customer’s problem. I don’t mean high-level problems. I mean the details of those problems that distinguish whether a solution is interesting or whether a solution is really disruptive.

2 . Understand the value of innovation. The value of an innovation is bounded from above by the salary you would pay a human to perform a similar task. If you’re coming up with an innovation that doesn’t replace a human or doesn’t do something better than a human can, the commercial value is limited. We get enamored with our research. We love our research and the technology. And we love solving hard problems. But just because you’ve solved the hard problem, if it doesn’t complete a task that is hard or impossible for humans, then it’s much easier to pay a human to do it. Also, just because a problem is hard, it doesn’t mean the solution is commercially viable.

3 . Understand how things scale. Having access to a disruptive technology is one thing. Deploying it at scale is very different. It’s not only knowing the problem and that we have a solution. It’s how do I actually deploy that solution at scale?

4 . Understand political and social landscapes. In many cases, it is possible to develop a great technology that solves a customer problem, and we understand how to deploy it at scale. However, sometimes we misinterpret how people want to use that solution. One timely example is the recently announced Apple Vision Pro. It’s a technological marvel and has the potential to do an enormous amount of work, but will wearing a mask for hours at a time ever become socially acceptable?

5 . Understand the Hype. Right now, ChatGPT has a tremendous amount of hype surrounding it. Everyone already believes this type of AI will change the world. That may or may not happen. With potentially disruptive technology, it is critically important not to over-hype the technology or the disruption. Successful technologies often get introduced multiple times, and each time it gets introduced, it doesn’t necessarily solve all of the problems it was expected to solve. Each time this happens, the disruption loses credibility. With ChatGPT, we’re already experiencing the negative effects of too much hype. It’s sucking the oxygen out of the story and will probably set back overall AI development by at least five years.

We are sure you aren’t done. How are you going to shake things up next?

I don’t know! And I think that’s the fun piece. I’m going to spend my time looking for interesting problems. We need to understand the core aspect of that problem and match it with interesting and innovative research.

And you can bet that the intersection of machine learning and science will create more disruptions that will change the world.

How can our readers follow you online?

www.nec-labs.com and on LinkedIn at: https://www.linkedin.com/in/christopher-white-0986421/

This was very inspiring. Thank you so much for joining us!